SEARCHSearch & retrieval
Jev vs. Embeddings for Video Indexing
Arun Agrahri@arunagrahri
1:06This project evaluates the effectiveness of Jev, in conjunction with embeddings, for pinpointing specific moments within a large video corpus.
I tested Jev against embeddings for finding the exact moment in a video for a corpus of over 3000+ hours of content in the health and wellness space.
> Alone: 67% each. Together: 75%. x.com/arunagrahri/status/21019488318816542
Related builds
Vectorless RAG with Jev
This project introduces a novel approach to Retrieval Augmented Generation (RAG) that replaces traditional embedding methods with a structured, table-of-contents-based navigation system, significantly improving speed and reducing costs by offloading non-generative tasks to Jev.
